Electronic and Optical Properties of Lead-Free

Hybrid Perovskite CH3NH3SnI3 from First Principles

Calculations

Ibrahim Omer Abdallah1,2, Daniel P Joubert1, and Mohammed S HSuleiman1,3

1 The National Institute for Theoretical Physics, School of Physics and Mandelstam Institutefor Theoretical Physics, University of the Witwatersrand, Johannesburg, Wits 2050, SouthAfrica.2 Department of Scientific Laboratories, Sudan University of Science and Technology,Khartoum, Sudan.3 Department of Basic Sciences, Imam Abdulrahman Bin Faisal University, P. 0. Box 1982,Dammam, KSA.

E-mail: ibraphysics@gmail.com

Abstract. Organic-inorganic halide perovskites have recently emerged as promisingcandidates for low cost, high-efficiency solar cells. In this work, the electronic and opticalproperties of the triclinic lead-free hybrid halide perovskite CH3NH3SnI3 as a solar cell absorberhas been investigated using density functional theory and many body perturbation theorycalculations. Depending on the functional used, our calculated band gaps are 0.82, 1.25 and1.32 eV, which agree well with the experimental result (1.21 and 1.35) eV. In addition, ourcalculations show that that CH3NH3SnI3 is a direct band gap semiconductor. Many bodyperturbation theory at the G0W0 level of approximation gives a fundamental band gap of 1.54eV. In order to obtain optical spectra, we carried out Bethe-Salpeter equation calculations on topof non-self-consistent G0W0 calculations. Our calculated optical band gap shows anisotropy withan absorption edge of 1.22 eV for out-of-plane polarisation and 1.25 eV for in-plane polarisation.These values lie within the experimentally reported range, confirming that CH3NH3SnI3 haspotential as a solar cell absorber.

1. IntroductionHybrid halide perovskites have recently emerged as potential new materials for solar cellapplications leading to a new class of hybrid semiconductor photovoltaic cells [1, 2]. It isexemplified by the remarkable increase in power conversion efficiency from 3.8% by Miyasaka [3]to over 20% by Korea Research Institute of Chemical Technology (KRICT) [4] using low costproduction methods. This performance is due to the exceptional properties of hybrid halideperovskites displaying high absorption coefficients, high carrier mobility, direct and tunableband gaps [5] and long charge carrier diffusion lengths [6, 7, 8]. Progress in efficiency, however, ishindered by many challenges such as the instability of many perovskite phases [9] and the toxicityof Pb in lead halide perovskites, the absorber in the prototypical high efficiency perovskite basedsolar cells [10]. Therefore, it is necessary to find an alternative material that does not containtoxic lead. A possible candidate is tin organic-inorganic halide perovskite, CH3NH3SnX3 (X=

Proceedings of SAIP2018

SA Institute of Physics ISBN: 978-0-620-85406-1 2

Cl, Br or I), which has been reported to have a band gap of 1.21 eV and 1.35 eV dependingon the preparation method [11]. Depending on temperature, CH3NH3SnX3 (X= Cl, Br or I)reveals a very rich phase diagram; i.e., the crystal structure of tin halide perovskite goes fromcubic, tetragonal, orthorhombic and monoclinic to triclinic phase by cooling [12]. At roomtemperature, CH3NH3SnI3 structure presents a cubic phase, and it changes to tetragonal at 275K and to orthorhombic at 110 K [13, 14], but there is no clearly define temperature range in theliterature for the triclinic phase.

The first lead-free solar cell made of tin halide perovskite was demonstrated by the Noel etal. group [10] and the Hao et al. group [15] in 2014. Hao et al. investigated the photovoltaicproperties of the tetragonal phase of CH3NH3SnI3 and found that it has a direct-gap with anenergy gap of 1.3 eV, which is significantly redshifted compared with CH3NH3PbI3, whose bandgap is 1.55 eV [16, 17]. Recent studies reported that CH3NH3SnI3 could serve as a lead-freelight absorbing material, and solar cell power conversion efficiencies (PCE) in the range of 5-6%have been obtained, however the stability of CH3NH3SnI3 remains a challenge [10, 15]. Sn-basedperovskites in particular, have shown excellent mobilities in transistors [18]. Theoretical studieswere carried out by Paolo Umari et al. [19] using GW (where G is the Green’s function and Wis the screened Coulomb interaction) calculations including spin orbit coupling of CH3NH3XI3(X = Pb, Sn) in the tetragonal phase. Their calculations gave band gaps of 1.67 eV and 1.10eV for CH3NH3PbI3 and CH3NH3SnI3 respectively. They showed CH3NH3SnI3 to be a bettercandidate for electron transport than CH3NH3PbI3.

In this work, we perform density functional theory (DFT) calculations of the electronicand optical properties focusing on the triclinic phase of the CH3NH3SnI3. To the best of ourknowledge, the electronic and optical properties of the triclinic phase of CH3NH3SnI3 have notbeen investigated extensively. We hope that the present investigation will contribute to a bettertheoretical understanding of the properties of this material, especially its potential in low bandgap solar cell applications.

2. MethodologyThe crystal structure of the triclinic phase of the hybrid halide perovskite (CH3NH3SnI3) isshown in Figure 1. The unit cell of CH3NH3SnI3 contains 24 H atoms, 4 Pb atoms, 4 C atoms,12 I atoms, and 4 N atoms, with a = 9.06 A, b = 9.06 A and c = 12.56 A (Material Project ID:mp-995238). In this work, the investigation of the electronic structure properties was performed

Figure 1. (Color online)The crystal structure of the triclinic phase of CH3NH3SnI3 (green:hydrogen; grey:nitrogen; yellow:carbon; red: iodine; blue:tin).

using the Vienna Ab-initio Simulation Package (VASP) [20, 21] based on Density FunctionalTheory (DFT) [22, 23]. The Projector-Augmented Wave (PAW) [24] method was employed to

Proceedings of SAIP2018

SA Institute of Physics ISBN: 978-0-620-85406-1 3

treat the electron-ion interactions. To describe the electrons exchange and correlation effects, weused the Generalized Gradient Approximation (GGA) as parametrized by Perdew, Burke andErnzerhof (PBE) [25], the modified Becke-Johnson (MBJ) [26] and the hybrid functional HSE06[27], where the Hartree-Fock screening parameter µ is set at 0.2 A−1. 4× 4× 4 Monkhorst-Packmeshes were used in sampling the Brillouin zone (BZ) with an energy cut-off of 520 eV. Theseparameters were found to be sufficient for energy convergence. The BZ sampling was chosenin such a way that the convergence of free energy is less than 1 meV/atom. The convergencethreshold for self-consistent field iteration was set at 10−6 eV. In order to study the opticalproperties, we performed non-selfconsistent G0W0 [28] calculations with 4008 bands and withan energy plane wave cut-off of 300 eV and a cut-off of 150 eV for the plane wave basis of theresponse function expansion. Optical spectra were calculated at the Bethe-Salpeter Equation(BSE) [29] level of approximation with input from a G0W0 calculations.

3. Results and Discussions3.1. Electronic PropertiesThe Kohn-Sham band structure of the studied material was calculated using the PBE functional.The calculated electronic band structure and its corresponding total and partial density ofstates (TDOS and PDOS) are displayed in Figure 2. The electronic band structure calculationsdemonstrate that this phase has a DFT direct band gap at the gamma point. We note fromthe PDOS that the valence band edge is dominated by I orbitals and the conduction band edgeby Sn orbitals. The organic cation CH3NH+

3 do not have any significant contribution aroundthe band edge. Our calculated approximate DFT and many body perturbation theory at theG0W0 level of approximation fundamental band gaps for the triclinic perovskite CH3NH3SnI3are listed in Table 1. The only available experimental data is that reported by Stoumpos et al.[11], where, depending on the preparation method, they reported that the experimental bandgap of tin based hybrid halide perovskites to be 1.21 eV and 1.35 eV. Our results are consistentwith these results.

X Γ Y|L Γ

Wave Vector

-10

-5

0

5

10

E-E

f (

eV)

0 30 60 90 120

TDOS (arb. unit)

TDOS

CH3NH

3

+

0 5 10 15 20 25 30

PDOS (arb. unit)

Sn p

Sn sI p

Figure 2. DFT calculated electronic structure of the triclinic phase of CH3NH3SnI3 using PBE:band structure (left), total density of states (TDOS) and the partial density of states (PDOS)due to CH3NH+

3 (middle); and PDOS due to Sn and I (right).

3.2. Optical PropertiesThe optical properties can be calculated from the complex dielectric tensor, ε(ω) = ε1(ω)+iε2(ω)which describes the polarization response of a material to an externally applied electric field E.Using the real part ε1(ω) and the imaginary part ε2(ω) of the dielectric tensor ε(ω), we compute

Proceedings of SAIP2018

SA Institute of Physics ISBN: 978-0-620-85406-1 4

Table 1. Our calculated and the experimental band gap of triclinic CH3NH3SnI3.

Functional PBE MBJ HSE06 GW Exp.[11]

Band gap (eV) 0.82 1.25 1.32 1.54 1.21 and 1.35

0 1 2 3 4ω(eV)

0

1

2

3

4

5

6

7

α(ω

)x10

5 cm-1

αxx

αyy

αzz

Figure 3. The optical absorption coefficient spectra of triclinic CH3NH3SnI3.

Figure 4. Tauc plot of the absorption coefficient, showing the polarization-dependent onsets.

the optical absorption coefficient spectra using the following equation.

α(ω) =√

2ω

([ε2re(ω) + ε2im(ω)]

12 − εre(ω)

) 12

. (1)

The results in Figure 3 show that the highest absorption peak in the region (1.0-2.5) eV ofαyy(ω) component is higher than that of αxx(ω) and αzz(ω). Figure 4 shows the absorptioncoefficient α(ω) for triclinic CH3NH3SnI3 plotted in Tauc formalism. From Figure 4 we cansee there are featured edges as a function of polarization. Due to the structural anisotropy weobserved variation of band edge in the different polarizations. For in plane polarisation the onsetis at 1.25 eV and for out of plane polarisation the onset is at 1.22 eV within the experimentallyreported values (1.21 and 1.35) eV.

Proceedings of SAIP2018

SA Institute of Physics ISBN: 978-0-620-85406-1 5

4. ConclusionsThe electronic and optical properties of the triclinic CH3NH3SnI3 have been calculated usingfirst-principles methods. The material was found to be a direct band gap semiconductor.Depending on the method used, our calculated band gaps are 0.82 eV (PBE), 1.25 eV (MBJ),1.32 eV (HSE06) and 1.54 eV (G0W0). Obtained values agree well with the experimental results(1.21 eV and 1.35 eV). Moreover, the BSE absorption edges using the Tauc form are found tobe 1.22 eV for out-of-plane polarisation and 1.25 eV for in-plane polarisation. This anisotropyseen in the BSE absorption edges may be related to the experimental result 1.21 and 1.35 eV.Inspection of the obtained band structure, PDOS and TDOS reveals that the organic cationCH3NH+

3 does not have significant contribution around the band edges as I and Sn. Our resultsconfirm that the triclinic phase of CH3NH3SnI3 has potential as a solar cell absorber. Finally,with the help of the real part εre(ω) and the imaginary part εim(ω) of the dielectric tensor ε(ω)we derive all the desired frequency dependent optical spectra such as absorption coefficient atBSE level of approximation, which show a significant optical anisotropy.

5. AcknowledgementIOAA would like to acknowledge the support he received from NRF-TWAS for funding, andSudan University of Science and Technology (SUST). We also wish to acknowledge the Centrefor High Performance Computing (CHPC), South Africa, for providing us with computingfacilities.

References[1] Green M A and Ho-Baillie A 2017 ACS Energy Letters 2 822–830

[2] Park N G 2013 The Journal of Physical Chemistry Letters 4 2423–2429

[3] Kojima A, Teshima K, Shirai Y and Miyasaka T 2009 Journal of the American ChemicalSociety 131 6050–6051

[4] Yang W S, Noh J H, Jeon N J, Kim Y C, Ryu S, Seo J and Seok S I 2015 Science 3481234–1237

[5] Noh J H, Im S H, Heo J H, Mandal T N and Seok S I 2013 Nano Letters 13 1764–1769

[6] Stranks S D, Eperon G E, Grancini G, Menelaou C, Alcocer M J, Leijtens T, Herz L M,Petrozza A and Snaith H J 2013 Science 342 341–344

[7] Dong Q, Fang Y, Shao Y, Mulligan P, Qiu J, Cao L and Huang J 2015 Science 347 967–970

[8] Shi D, Adinolfi V, Comin R, Yuan M, Alarousu E, Buin A, Chen Y, Hoogland S,Rothenberger A, Katsiev K et al. 2015 Science 347 519–522

[9] Zhang Y Y, Chen S, Xu P, Xiang H, Gong X G, Walsh A and Wei S H 2018 Chinese PhysicsLetters 35 036104

[10] Noel N K, Stranks S D, Abate A, Wehrenfennig C, Guarnera S, Haghighirad A A, SadhanalaA, Eperon G E, Pathak S K, Johnston M B et al. 2014 Energy & Environmental Science 73061–3068

[11] Stoumpos C C, Malliakas C D and Kanatzidis M G 2013 Inorganic Chemistry 52 9019–9038

[12] Feng J and Xiao B 2014 The Journal of Physical Chemistry C 118 19655–19660

[13] Parrott E S, Milot R L, Stergiopoulos T, Snaith H J, Johnston M B and Herz L M 2016The Journal of Physical Chemistry Letters 7 1321–1326

[14] Takahashi Y, Obara R, Lin Z Z, Takahashi Y, Naito T, Inabe T, Ishibashi S and TerakuraK 2011 Dalton Transactions 40 5563–5568

[15] Hao F, Stoumpos C C, Cao D H, Chang R P and Kanatzidis M G 2014 Nature Photonics8 489

Proceedings of SAIP2018

SA Institute of Physics ISBN: 978-0-620-85406-1 6

[16] Eperon G E, Stranks S D, Menelaou C, Johnston M B, Herz L M and Snaith H J 2014Energy & Environmental Science 7 982–988

[17] Ali I O A, Joubert D P and Suleiman M S H 2018 Materials Today: Proceedings 5 10570–10576

[18] Kagan C, Mitzi D and Dimitrakopoulos C 1999 Science 286 945–947

[19] Umari P, Mosconi E and De Angelis F 2014 Scientific Reports 4 4467

[20] Kresse G and Hafner J 1993 Physical Review B 47 558

[21] Kresse G and Hafner J 1994 Physical Review B 49 14251

[22] Hohenberg P and Kohn W 1964 Physical Review 136 B864

[23] Kohn W and Sham L J 1965 Physical Review 140 A1133

[24] Kresse G and Joubert D 1999 Physical Review B 59 1758

[25] Perdew J P, Burke K and Ernzerhof M 1996 Physical Review Letters 77 3865

[26] Becke A D and Johnson E R 2006 The Journal of Chemical Physics 124 221101

[27] Krukau A V, Vydrov O A, Izmaylov A F and Scuseria G E 2006 The Journal of ChemicalPhysics 125 224106

[28] Hedin L 1965 Physical Review 139 A796

[29] Salpeter E E and Bethe H A 1951 Physical Review 84 1232

Proceedings of SAIP2018

SA Institute of Physics ISBN: 978-0-620-85406-1 7